Introduction: Treatment options for children with HIV-1 are limited. We aimed to compare activity and safety of three dual-nucleoside analogue reverse-transcriptase inhibitor (NRTI) regimens with or without a protease inhibitor in previously untreated children with HIV-1.
Methods: In our multicentre trial, we randomly assigned 36 children to zidovudine and lamivudine, 45 to zidovudine and abacavir, and 47 to lamivudine and abacavir. Children who were symptom-free (n=55) were also randomly assigned to receive nelfinavir or placebo. Children with more advanced disease received open-label nelfinavir (73). Primary endpoints were change in plasma HIV-1 RNA at 24 and 48 weeks for the NRTI comparison and occurrence of serious adverse events for both randomised comparisons. Analyses were by intention to treat.
Findings: Children had a median CD4 percentage of 22% (IQR 15-29) and a mean HIV-1 RNA concentration of 5.0 log copies/mL (SD 0.8). One child was lost to follow-up and one died of sepsis. At 48 weeks, in the zidovudine/lamivudine, zidovudine/abacavir, and lamivudine/abacavir groups, mean HIV-1 RNA had decreased by 1.71, 2.19, and 2.63 log copies/mL, respectively (estimated in absence of nelfinavir) (p=0.02 after adjustment for baseline factors). One child had a hypersensitivity reaction to abacavir; and three with possible reactions stopped abacavir. There were 24 serious adverse events--six in the symptom-free children (all on nelfinavir), but none were attributed to nelfinavir.
Interpretation: Regimens containing abacavir were more effective than zidovudine/lamivudine. Such regimens could be combined with protease inhibitors and non-nucleoside reverse transcriptase inhibitors for safe and effective treatment of previously untreated children with HIV-1.